BBD-MN3010 and MN3011- is a MN3101 clock necessary?

[Mike I.]

Thanks for elaborating on the post from last week Rene! I was curious 
about how the Panasonic BBD clocks behaved, and finally was able to look 
at them with a scope, seeing that they are indeed non-overlapping. My 
interpretation of your original post was that if an overlapping clock is 
used, make the overlap time as short as possible to minimize leakage 
between adjacent cells. Since the two clock phases are  ideally mirror 
images of each other this would cancel the capacitive clock feedthrough 
at the outputs. This would require a very fast, symmetrical, 
low-impedance driver circuit. Sampling in the middle of the pulse is an 
interesting idea - resembles what is done in asynchronous serial data 
links. Mike I.